This report aims to address the debate over whether it is more beneficial to hire new employees when faced with shortages, or cover the shifts with overtime. The issue is examined from the viewpoint of Columbia University’s Department of Public Safety, and takes into account factors and costs as found therein. This report includes a graphical comparison of the costs associated with each option, and analyses and discusses the greater expense of hiring.
One of the most debated topics in management is whether it is more beneficial to hire new employees or pay overtime to increase production or coverage. Many factors go into this decision, including benefits, duration of the period of extra work needed, and even type of work involved. In Columbia University’s Department of Public Safety, this issue gets more attention than at most places of business. Minimal staffing levels, union regulations, and round-the-clock posts – among other factors – conspire to make this decision one that is faced often. When all factors are considered, is it better for the department to hire new officers when coverage is lacking, or to utilize overtime to staff positions as needed?
In almost all instances, it is more cost-efficient to fill gaps in coverage with overtime rather than hiring new officers in this department. This report will examine the factors involved, analyze the financial implications, and discuss exceptions to the findings. Explicit and implicit costs to the department will be discussed, along with their relevance to the decision-making process. This report will also compare the costs of new hires and overtime to find an equilibrium point beyond which the decision should change, and introduce the concept of diminishing returns. Finally, it will summarize the process and discuss application.
In Columbia University’s Department of Public Safety, there is a minimum level of on-duty staffing required at all times. Department employees are licensed, uniformed personnel belonging to the Transit Workers’ Union, Local 241. Management is required to ensure sufficient coverage on a daily basis, while conforming to the guidelines of the collective bargaining agreement. Sick time, vacation time, requests for guard services or special details, and emergencies such as natural disasters can create shortages in coverage that need to be addressed. Failure to cover these openings is not an option, and so management is left with two choices: hire additional full-time personnel to cover the openings, or pay overtime – at the rate of one-and-one-half times normal salary. Spending large amounts on overtime raises concerns in multiple areas, such as yearly budget reviews and union negotiations. This report aims to examine the issue in depth, and decide which option is more fiscally appropriate in a given situation.
Factors and Costs
As previously mentioned, there are numerous factors that influence this decision. The following are the most critical points that must be considered: * Department employees are unionized. Due to contractual agreements, the department must maintain all currently occupied positions. Any vacated position must be filled, somewhere in the department; for example, if an officer resigns, the department must fill that vacancy, whether in the same position or laterally, or be found in violation of the collective bargaining agreement. * Note that this is regardless of shortages, and only applies to permanently vacated positions. If the department has three extra officers on a given tour, and one resigns, the department may not continue to operate with two extras; rather, it is necessary to hire a new third extra. If an officer goes out on extended disability, however, it is not contractually necessary to fill that opening. In these cases, the department can cover the shortages, if any, through other means.
Hiring remains an option, but the mandate to retain positions applies to newly created positions as well. In effect, if a new officer is hired in this situation, and the disabled officer returns to duty, the department is now obligated to keep both positions filled permanently and has a new minimum level of staffing. * New hires are paid at a rate of 80% salary for the first two years, then 82.5% the third year, 85% the fourth year, then 100% from year five on. * Overtime is paid at one-and-one-half times salary for any work done beyond 40 hours in a week. If an employee works a shift during all 7 days of the pay period, work done on the 7th day is paid at twice salary. * Any mandatory overtime performed without at least one hour of advance notice is paid at double time. This is relevant because when electing to not cover openings with additional personnel, any emergency shortages could lead to mandatory overtime without notice.
* Most posts on campus require continuous coverage 24 hours a day, 365 days a year. This means there are three distinct tours with their own unique dynamics for coverage. The 8×4 tour, for example, has the highest concentration of senior officers, meaning they accrue more vacation time than junior officers. The summer months are obviously prime time for vacation picks, and two officers are allowed out at a time during summer. This means that summer vacation time is booked solid on the 8×4, so the tour is perennially short two officers all summer. Also, as a medical research facility in addition to a university campus, there are more posts open during the 8×4 tour than any other. As a result, it makes a difference where the shortage occurs. * Shortages on the night tours are easier to absorb, because there are fewer posts, but there is also night differential to consider: any officer working between 4pm and 8am makes an extra stipend of $1.80 per hour for night differential. As a fixed bonus separate from salary, however, night differential has no impact on this decision, as shown later.
* New employees must be trained for a minimum period of one month before being able to assume post unsupervised and unassisted. This not only reduces the utility of a new hire for the first month, but also reduces the utility of the officers assigned to train him or her. * The department provides, as a service, officers for hire on campus for special events. The department charges $50 hourly for this service per officer assigned. If there is extra coverage on the tour during which the guard service is requested, an officer may be reassigned from a non-mandatory post, at a significant savings to the department. By using already assigned personnel as opposed to paying overtime for another officer in addition to the scheduled non-mandatory officer, the department can significantly reduce the overtime budget.
For the purposes of this report, it will be assumed that no overtime is covered in this way, to preserve an empirical comparison of salary costs. * Officers are paid hourly, and thus are a variable labor cost. If the amount of time worked during a pay period changes, the officer’s salary for that period will change accordingly. For the purposes of this report, it will be assumed that labor costs are fixed. All officers not assigned to overtime always work exactly 40 hours per week, and those assigned to overtime always work in increments of exactly 8 hours per additional shift. Docking pay for tardiness will be disregarded.
* The department pays an amount equal to 37.7% of each officer’s salary for his or her benefits package. This payment remains in effect so long as the officer is on the University’s payroll, whether or not he or she actually works or is paid. Officers on vacation, sick leave, or disability still incur this cost. While the cost of benefits itself is highly relevant, and will be used extensively in analysis, the fact that disabled officers still incur this cost even when not being paid by the department is not relevant to this discussion.
In trying to determine how best to fill coverage, the most obvious comparison is between costs of a new hire and overtime. As previously stated, overtime is equal to one-and-one-half times salary (S). Cost of overtime can therefore be represented by the equation COT=h(1.5S), where h represents hours worked. New hires, who incur a cost of 37.7% of their salary, have a cost of CNH=h(1.377S). New hires always have a starting salary of 80% of full pay, while officers working longer than two years have a gradually increasing pay scale. One eight hour shift of salary at the rookie wage scale is equal to $149.20. For a third-year officer, that rises to $153.84, then $158.56 for a fourth-year employee. Finally, upon completion of four years of employment, the wage scale tops out at $186.48.
A new hire will always work 5 shifts a week, leading to CNH=5(1.377)(149.2), or CNH=1027.24. COT, on the other hand, varies based on how many shifts of overtime are required that week, and the seniority of the officer performing the overtime. Plotting those equations against each other leaves us with the first graph below, labeled Weekly Salaries, New Hires vs Overtime at Each Level of Seniority. As expected, officers receiving full pay cost significantly more to cover with overtime than any other officer, but even at that escalated rate, there would need to be shortages in excess of 3 positions per week to justify hiring a new officer. For all other levels of seniority, 4 shifts of overtime could be covered per week at a lower cost than one new hire. In order to simplify the situation, this report will henceforth take an average of the salary levels and assume that all officers working overtime earn that rate of pay.
Weekly Salaries, New Hires vs Overtime at Each Level of Seniority After applying this assumption, full pay is removed as an outlier, and it becomes always less expensive to fill four shifts of overtime than to hire one new officer.
Weekly Salaries, New Hires vs Average Overtime Rate for Day Shift As seen in these graphs, the presence of night differential increases cost equally across the board, and therefore has no impact on this decision.
Weekly Salaries, New Hires vs Average Overtime Rate for Night Shifts
The graphs provide striking evidence that new hires are significantly more costly for small to reasonable shortages, seemingly four or fewer per week. However, due to the nature of the department, this is an underestimation. Since all three tours need to be covered, and officers must contractually be assigned to a stable tour, each tour must have its own graph. It is clear in every graph above that if there are five or more shifts of overtime per week, it is always less costly to hire a new officer; this is not completely accurate. During a given stretch, if there were expected to be 6 shifts of overtime that need to be filled each week, but the shifts are distributed evenly, with two shortages per tour per week, hiring one new officer will not account for five tours of overtime, but two.
As presented by Bob Thomas (2006, p.13), “adding officers does not automatically reduce an equivalent number of overtime hours by the amount of the actual hours worked.” Thomas explains that while adding officers reduces overtime, it also raises costs disproportionately in an example of diminishing returns, an effect found mirrored in the previous graphs in this report. In essence, hiring additional officers beyond a certain point will produce reduced utility with each new hire. Hiring additional officers will add utility, but at a much higher cost per unit of utility, creating an inefficient Summary
The measurements set forth in this report clearly show that only the most extreme of shortages justifies hiring an additional employee from a cost-effectiveness standpoint, but this is not the only factor to consider. Most important among these is the contractual obligation of the department to hire a new officer each time a position is vacated. Fatigue is a concern that is often raised regarding the use of overtime to fill gaps in coverage. Steve Earley’s report for the Riverside Fire Department (2001, p.14) cited studies finding a correlation between the number of hours worked and work-related injuries.
The department has already taken steps to mitigate this, as employees are prohibited from working longer than 16 consecutive hours, and are strongly dissuaded from working 7-day weeks. Overtime is also distributed in a rotation, by seniority, so that no one officer has significantly more overtime opportunities than any other. In conclusion, the data is so compelling that whenever possible, shortages should be covered with overtime rather than hiring.
Earley, S. (2001). An Analysis of the Utilization of Overtime Versus Hiring Additional Personnel. Riverside, CA. Thomas, B. (2006). Corrections Overtime Planning Study. Olympia, WA.